https://doi.org/10.65770/LIKR1073
ABSTRACT
Gold futures prices are dynamic financial instruments that are often influenced by economic uncertainty, market changes, and investment sentiment. This study aims to compare the accuracy of Autoregressive Integrated Moving Average (ARIMA) and Least Square Spline in forecasting daily gold futures prices. The data used in this study consist of daily Gold Futures prices from 4 January 2021 to 30 December 2025, totaling 1063 observations, which were divided into 850 training data and 213 testing data. The ARIMA model was developed through stationarity testing, differencing, ACF-PACF identification, model selection based on AIC, and residual diagnostics. The Least Square Spline model was developed by selecting the spline order and the optimal number of knots based on Generalized Cross Validation (GCV). Accuracy was evaluated using RMSE, MAE, and MAPE. The results show that the best ARIMA model used was ARIMA(1,1,0), while the best Least Square Spline model was a first-order spline with 15 knot points. Based on the evaluation results on the testing data, ARIMA(1,1,0) produced an RMSE of 909.5807, an MAE of 761.1346, and a MAPE of 20.59894%. Meanwhile, Least Square Spline produced an RMSE of 355.8918, an MAE of 267.4223, and a MAPE of 7.11823%. Therefore, Least Square Spline provides better prediction accuracy than ARIMA in forecasting daily gold futures prices.
References
[1] S. Setyowibowo, M. As’ad, Sujito, and En. Farida, “Forecasting of Daily Gold Price using ARIMA-GARCH Hybrid Model,” J. Ekon. Pembang., vol. 19, no. 2, pp. 257–270, 2021, doi: 10.29259/jep.v19i2.13903.
[2] Y. R. Madhika, Kusrini, and T. Hidayat, “Gold Price Prediction Using the ARIMA and LSTM Models,” Sink. J. dan Penelit. Tek. Inform., vol. 7, no. 3, pp. 1255–1264, 2023, doi: 10.33395/sinkron.v8i3.12461.
[3] F. Yuan, C. Lee, and C. Chiu, “Using Market Sentiment Analysis and Genetic Algorithm-Based Least Squares Support Vector Regression to Predict Gold Prices,” Int. J. Comput. Intell. Syst., vol. 13, no. 1, pp. 234–246, 2020, doi: 10.2991/ijcis.d.200214.002.
[4] H. R. Sari, S. Wahyuningsih, and M. Siringoringo, “Forecasting Indonesian gold prices using the ARIMA(0,1,1)-GARCH(1,0) model,” J. EKSPONENSIAL, vol. 15, no. 2017, pp. 1–10, 2024, doi: 10.30872/eksponensial.v15i1.1265.
[5] T. Bunnag, “The Importance of Gold’s Effect on Investment and Predicting the World Gold Price Using the ARIMA and ARIMA-GARCH Model,” Ekon. J. Econ., vol. 2, no. 1, pp. 38–52, 2024, doi: 10.60084/eje.v2i1.155.
[6] D. Makala and Z. Li, “Prediction of gold price with ARIMA and SVM,” J. Phys. Conf. Ser., vol. 1767, no. 2021, pp. 1–8, 2021, doi: 10.1088/1742-6596/1767/1/012022.
[7] G. S. Pillai and M. I. Mary, “Gold Price Forecast Based on the Least Square Support Vector Machine,” Int. J. Intell. Syst. Appl. Eng., vol. 12, no. 3, pp. 684–697, 2024.
[8] M. Yurtsever, “Gold Price Forecasting Using LSTM , Bi-LSTM and GRU,” Eur. J. Sci. Technol., vol. 31, no. 31, pp. 341–347, 2021, doi: 10.31590/ejosat.959405.
[9] Z. Amry and B. H. Siregar, “ARIMA Model Selection for Composite Stock Price Index in Indonesia Stock Exchange,” Int. J. Account. Financ. Stud., vol. 2, no. 1, pp. 31–38, 2019, doi: 10.22158/ijafs.v2n1p31.
[10] A. Sofian, E. Hidayat, and G. Primajati, “Estimating and Forecasting Composite Index in Pandemic Era Using ARIMA-GARCH Model,” J. Varian, vol. 7, no. 2, pp. 115–124, 2024, doi: 10.30812/varian.v7i2.2103.
[11] A. T. Fitriyah, N. Chamidah, and T. Saifudin, “Comparison Of Least Square Spline and ARIMA Models For Predicting Indonesia Composite Index,” BAREKENG J. Math. Its Appl., vol. 19, no. 3, pp. 2169–2178, 2025, doi: 10.30598/barekengvol19iss3pp2169-2178.
[12] A. Araveeporn, “The Estimating Parameter and Number of Knots for Nonparametric Regression Methods in Modelling Time Series Data,” Modelling, vol. 5, pp. 1413–1434, 2024, doi: 10.3390/modelling5040073.
[13] C. Andreas, F. Y. H, E. J. S, and N. Chamidah, “Analyzing The Effect of BI 7-Days Repo Rate on The Jakarta Composite Index Using Nonparametric Regression Approaches Based on Least Square Spline Estimator Berdasarkan Estimator Least Square Spline,” J. Mat. Stat. Komputasi, vol. 17, no. 3, pp. 447–461, 2021, doi: 10.20956/j.v17i3.13101.
[14] A. T. Fitriyah, N. Chamidah, T. Saifudin, B. Lestari, and D. Aydin, “Least Squares Spline Estimation Method in Semiparametric Time Series Regression for Predicting Indonesia Composite Index,” Stat. Optim. Inf. Comput., vol. 14, pp. 3776–3803, 2025, doi: 10.19139/soic-2310-5070-2704.
Download all article in PDF
